We helped plan and implement a study to estimate the risk of breast cancer from BRCA1 and BRCA2 mutations. Blood samples were obtained from over 5,000 Ashkenazi men and women in the Washington, D.C. area. These people are Jews whose ancestors lived in Central and Eastern Europe. The study will estimate the excess risk associated with mutations by comparing the cancer histories of relatives of study subjects who carry a mutation with the histories of relatives of non-carriers. This study will also give an estimate of the prevalence of these mutations in the Ashkenazi population. Methodologic studies are underway to devise efficient designs and methods of analysis for studies of this type. We discussed the use of genetic markers, including somatic mutations and inherited polymorphisms, to plan and analyze intervention trials. Efficient trial designs can take advantage of markers to reduce response heterogeneity, to identify subgroups at higher risk of cancer, and to identify subgroups likely to benefit from treatment. One collaborative study defined the clinical features of resistance to thyroid hormone by comparing affected members in 42 kindreds with unaffected relatives. Inheritance was autosomal dominant in 22 families, sporadic in 14 families, and unknown in 6 families. Those with thyroid hormone resistance had increased incidence of goiter (65%), attention-deficit hyperactivity (60%), IQ below 85 (38%), and speech impediments (35%) compared to controls. We studied the role of Epstein-Barr Virus (EBV) in familial Hodgkin's disease (FHD). There was no evidence of increased concordance (assessed by in situ hybridization) of EBV RNA among FHD cases who were first degree relatives, nor was there a difference in seroprevalence of antibody to EBV viral capsid antigen among cases compared to controls. These data suggest that EBV does not play an important role in FHD. Risk of renal cell carcinoma was found to be associated with hypertension and with the use of diuretics and other antihypertensive drugs. The study was unable to disentangle the effects of hypertension from those of medication, however, because these factors were subject to misclassification error and were highly correlated.